E61: Alex Svanevik: Finding Crypto Alphas Before Everyone Else!

43:12I don't think you should do that every year, actually. But we had to do it for many different reasons. So the whole thing is built from the ground up, which first of all means we can ship new things faster, which is needed in crypto. You need to move very fast, and speed is literally one of our values, although I don't think we were living up to it that well in 2022. I think that's now changing. the other part was we literally have made the product itself like a hundred times faster so if you're a user and you just want to get the answer to your question both in terms of the navigation you can just smash command k and write you know little pudgies or whatever you want and you're taken to the right dashboard you don't have to like do i go to nft god mode do i go to wallet profiler we just know where you want to go based on your input okay it sounds kind of obvious but that was not the case in one so the path is much shorter from question to answer and then the literal performance of the database that powers these queries is has is a different technology and so it is a hundred times faster to just get the loading of whatever you're looking at which is a huge benefit and then we've also added several i think very innovative features that make the whole product more personalized and makes it feel more streamlined for example so we have this feature called smart segments where you know historically people know nonsense for our ability to label wallets and this is still something we're the best at we have more than 300 million addresses labeled across lots of different chains and now though we have created this functionality where you can decide your own criteria like let's say you wanted to monitor everyone who has made i don't know 10 eth on board apes you can drag and drop like or select some filters very quickly and you configure that you know traded token board ape profit 10 eth boom and then it loads instantly a segment of wallets you can now save that segment and say board ape profiteers or something like that you can also do the opposite who lost money on board apes or whatever and we're adding more and more criteria for that and you can then create these different segments that you can first of all you can profile the segment so what other things do those wallets hold today right so that's kind of your own smart money or your own segment of wallets that you want to track and then you can bring those segments into the rest of the product and filter and set alerts and things like that for it so it just allows you to customize things much more but you code and it's super snappy to set it up so that's one example we also have a feature called signals where you don't have to because many people feel like they go to nonsense they need to have like a thesis they need to know what they're looking for and what you can do with signals it's literally a feed that just surfaces anomalous events on chain very smart right so hey this token has like 500 times the amount of inflow to centralized exchanges than it's ever had like what's going on there well let's take a look so we surface that for you and you can even filter within that signals feed do you care about nfts do you care about tokens you care about smart money do you care about exchanges like what do you care about and you can personalize that signals feed uh it's also ai driven which is cool and we're making more we're adding more ai features this uh this uh year which is pretty exciting and we we should talk about ai as well because i've spent a lot of time on it we will we will So there's a part we added to this podcast, which is called basically crypto alpha.

47:11Yeah. Because that's what people want on Twitter, you know, some concrete alpha. Yeah. And I think there is a good link between Nansen, Nansen2 and alpha. Yes. So the first question would be, why should people sign up for Nansen2? Yeah. And can you give some personal, you already did it last time, some personal examples of coins or NFTs that you managed to spot on early thanks to Nansen. Yeah. Yeah, it's funny. So when we started Nansen 2, when we launched it, I actually spun up a new wallet and funded it and said, let me just buy stuff I discover in Nansen 2 and see how it goes. and in about two weeks i had made more money on that portfolio actually i think i doubled the amount of money i needed to pay for a professional subscription for a year now that doesn't happen with everyone i can't guarantee that's gonna happen with everyone but uh some of the coins that I discovered through that were,

48:25I think, what were they? There was one, Cypher was one of them. I already knew about this one, but it came up in my feed because I saw other wallets that had been accumulating a position in the past buying even more. So that's one example where you can see on chain what people are actually doing. So people talk about stuff on Twitter, but you don't know how many of them are like bots or LARPers or whatever. On-chain, you see what people actually do. So yes, you can talk the talk, but do you walk the walk? That's what you see on-chain. So that is a fundamental principle, I think, is why people sign up for Nansen.

49:05The challenge with on-chain is that there's so much stuff going on. So you need some way to curate and surface the signal in all of it. And that's what we do. That's literally what the company has been created to do, to surface the signal. So Cypher is one of the coins. I knew about it. I was originally an angel investor in it. And after seeing what was happening on chain, I decided, and seeing, frankly, some of the top holders that were accumulating a position, because some of the funds, I don't know, you can actually see announced, and it's CMT Digital had accumulated like an 8 % of all the supply throughout the bear market for that coin.

49:51Yeah, so that's one example. There were a few other ones I looked at as well. RBX, I think, was one that I hadn't really heard of before. In hindsight, when I looked at Twitter, people had been talking about it. and because I saw smart money buying it I bought it and basically went up I think it went up 50 or 60 percent after that again this doesn't always happen there are no guarantees but these are examples of coins that kept coming up as I was using it and then if then there were some coins as well that I bought that went down and what I did was I you know sold the losers and I kept the winners which is I think the opposite of what many people do you tend to sell the winners because you want to take profits and you hope that the losers are going to come back which barely works which doesn't happen very often yeah but the reason people just sign up I mean if you don't know what's happening on chain and you're investing in crypto then I think you have a serious disadvantage it's not going to give you everything you need because there's more stuff happening around in the world that's you know valuable information than just on chain but it is a must-have like it's a necessity but it's not sufficient in terms of doing investing in crypto so i would say like on-chain and your social network are the two absolutely most important information sources and you can refine both of them you should actually spend time getting better at digesting information from both of those two channels the qualitative social network curate the people around you speak who you speak with who you take input from and then for on-chain get good at actually reading on-chain with the right tools right and there are people who use like ether scan quite uh proficiently uh it's not really built for alpha it's more like a reference product um but you know if you have the time to do that uh you can but what we've seen is when people invest say full time they tend to save a lot of time by using nonsense because of all the features that are built specifically around investing and because we have a particular focus on surfacing the signal not just to give you kind of a blockchain explorer where you can see everything that's going on let's continue with some crypto alpha um yeah let me just think about the i feel like i should always bring some alpha into a meeting but i never do and i have to like come up with on the spot with it uh i mean but cypher is probably the main one you even told me last time, and it was not Nansen too, but Nansen helped you get very early.

52:51I think it was, was it Bored Ape? Yeah. But then you, I mean, you had a lot. I sold all of them. And then you sold all of them, right? Yeah. Did Nansen help you find Pudgy Penguins? What's the story with Pudgy Penguins that made you, because you probably receive a million advisory roles, offers a day or a week, right? What's the story? What's your story with Pudgy Pengui? Because you were really early on. Yeah. And why did you decide to accept being an advisor? And what was really special about that kind of community and project back then? So the backstory on Pudgy is my friend Tom, who works at Dragonfly.

53:34We used to work together at Zero X in 2019. and so my friend Tom saw a tweet about a CryptoPunks billboard in San Francisco and he made this snarky tweet where he said, look, I'm just going to say it, no one's ever going to put pudgy penguins on a billboard. And I saw that and I was like, and then I saw the pudgy penguins and I thought they look kind of cute. And so I thought someone should put pudgy penguins on a billboard. and I ended up in a telegram group with some people who were Pudgy Penguins holders and they were hilarious frankly and I said look we should how do we make this happen and long story short some people in that group managed to get Pudgy Penguins on a billboard in KL in Malaysia in the center of the city during COVID so no one was there and advertising was very cheap but it made for a very good, you know, some good content in terms of like filming it and so on.

54:36And it was kind of a first community win for Pudgy Penguins, if you will. And then from there, I started seeing some parallels to other successful projects I had invested in or that I had, you know, spent time with. For example? Axie Infinity was the primary one. The parallels. Yes. And there were some parallels that actually, to me, were very obvious. And I was like, that I think now people are going to go back and say, yes, they were obvious, like given the price action, all that stuff that's been happening. But first of all, the IP is very cute. And I think cuteness is actually extremely valuable.

55:15It's like a very intrinsic and universal thing. So the IP is cute, both Axie and I think, I would argue Pudgee, even cuter. The second part is the community. and Axie I don't know if people know this but they had a very hardened community in say 2019 before everything you know started going super crazy and people were very engaged with that whole communities playing the economics of it the kind of organic scholars that were popping up people kind of sort of leasing out their Axies and other people playing the game with their Axies So there was this very strong community around it. And similarly, Pudgy Penguins was getting this very organic and strong community.

56:02And, you know, with everything that happened with Pudgy Penguins, that community proved to be very resilient. Is this pre-LUCA takeover? Yeah, exactly. So this was like, you know, you're seeing the signs of a strong community, but it wasn't fully formed and hadn't gone through that many trials yet. But then the whole thing happened with the rogs, right? They put fishing rods in our eggs. It's kind of the meme, right? There was like this egg. People don't know what's in it. Everyone thinks it's going to be a cute little pudgy, which I think came later. And then you open the egg and it's a fishing rod.

56:43And it made absolutely no sense. And I think that became like a crisis in the community in terms of the leadership. They thought, what the hell are these guys doing? This makes absolutely no sense. And in hindsight, it kind of became part of the lore because it was so funny and silly. But many people then stepped up and said that they wanted to buy out the whole project and take over the project. And the one who was successful in the end was Luca. And so this added a third C to, you know, we have Qt, we have Community, and I have a CEO as well. So there's a third C that comes in there. and frankly Luca is an incredible leader he is so inspiring he's a visionary and he's a real hustler like he's a he's kind of it's a bit unusual I think in the venture startup tech sector to see someone who has this kind of hustle energy I mean startup founders are often more kind of they are visionary they're technical you think in you know very big terms like you know high how high is up is kind of the sort of vc way of thinking about how big can this be but he is more like a scrappy hustler kind of guy and he has a track record of running businesses making lots of money selling them and so on and he's very young uh he also he also i think has very good values and has high integrity.

58:13But these things I kind of learned over time. I didn't know that immediately. But he is a big portion of why I have been and I'm still very bullish on Budget Penguins as a community, as a project. There is a fourth C2, which is the content. So this is the thing that he saw that as the CEO, I think was one of the great calls he made. It's like the content is really good. The IP is really strong. That's the thing that's super valuable. how do you just create more content and he and the rest of the team dominated giphy with pudgy penguin gifs and it's the most obvious play in the world in hindsight but they have like billions of views on giphy and people just use pudgy penguins gifs and without knowing it's freaking pudgy penguin exactly like literally on teams on whatsapp everywhere they don't need to know it's NFT.

59:05Why would you need to know it's an NFT? It's just a cute penguin. And then they also, of course, have the largest Instagram of any NFT collection, more than a million followers. So they really nailed the content piece. And by the way, the Instagram profile has a very different content strategy than most NFT collections would have, right? It's focused on mental health in a very positive way and so on and so forth. That's why it was so viral. Yeah. And so I think they also because he was so good at doing business and thinking about how do you get revenue and things like that they were able to roll out the toys and the plushies and all that stuff and i think he realized that like look you bootstrap a business from this community you don't extract the value from the community which is what many nft collections actually do right through like royalties or you know just selling more you know diluting your collection like the azuki's elementals is like the canonical example of that in my view you just dilute the the original holders and you actually dilute your community which i don't think is a good thing and so he saw this as an opportunity to go outside of the nft bubble into a much broader mainstream market and sell toys get the ip out there and do licensing deals which i think he's going to crush it at like over time builder build a real business essentially yeah or in nft like yeah it's not big thing right yes and so so i think like luca is the type of person who makes me very bullish on the gen z generation actually uh i think they like gen z they kind of get a lot of heat uh and they get ridiculed but there are some people who are just incredible founders like mr beasts you know as another example yeah just unbelievable entrepreneurs and so hard-working so smart so creative and uh anyway so that's kind of how i ended up um the the journey more or less on pudgy penguins and then he asked me to be an advisor because i'd you know supported the the project over time and you know helped him a bit with a fundraise uh and uh yeah it's amazing to see this you know this vision that that he had come to reality and to see all the work that people have put into that project come to fruition i usually don't like to talk about prices on this podcast because i like to do like sort of timeless non-chilling interviews but you tweeted I love so much you tweeted 10k per ETH 100 ETH per penguin ice cold mathematics which means I mean obviously you're a bit memeing here but like which means a million US dollars per penguin yeah what's your rationale if there is this is my like biology prediction right yeah no it's not very mathematically based I just think that they can reach that's i if you break down the two components do i think pudgy penguins can reach 100 yes yes absolutely do i believe eth can reach 10 000 yeah yeah absolutely it's just like simple sort of heuristics that makes sense you look at what uh bored apes or crypto punks previous cycle and like how pudgy is yeah it's yeah so it's like actually sometimes you you decompose something and it's like do these assumptions separately make sense yes and you put them together well it's the conclusion right that's why it's mathematics and so who knows if you'll get there again you know do your own research but what i do know is that the holders and the community of the poji penguins is so hardened it's insane it's so resilient it's gone through so much Shit.

1:02:57And... it's been attracting and you're probably a big part of that or at least a certain part of that the biggest community of builders in the space like that's right all for me one of the key things was like okay should i buy a pudgy and i mean at the same time did i influence you to buy a pudgy absolutely okay absolutely and and uh i think kevin from gojek he told me ah alex told me to buy a pudgy and i'm like the co-founder of gojek buys a pudgy alex has a pudgy then i'm like i look at a Layer Zero co-founder, a bunch of people are like pudgy, pudgy, pudgy, but it's kind of like, right? Igloominati.

1:03:37Like, absolutely, Igloominati. So basically, then I'm like, hmm, interesting. Then Luca comes on the podcast. I'm like, fuck man, the guy is really next level. I mean, you get very bullish when you meet him, right? He is very inspiring. And then, and then you need to Bobby from CoinGecko. I mean, basically, and now everyone, now everyone is going crazy on that. But even before, if you're on the lookout, six months ago, eight months ago, The key builders in the space, Pudgy Penguin PFP. So it was all there in plain sight, right? That's right. How, except, I mean, maybe with some of your help, but how did this happen that most of the biggest builders, which means, by the way, these people, first they're builders, so they're going to support the founders.

1:04:20Second, they have money, so they're less likely to sell, right? Yeah. How is it possible that the community is so strong and attracted everyone? A lot of it. I mean, it comes down to some of those fundamentals, like I was saying before, you know, the QtIP, the community and so on. But you also just have to hustle, you know, like the stuff we just talked about now. Like you handpick Apuji for the Deal CEO. I don't know if you know Deal. It's the fastest growing SaaS company in history. It's basically like an HR platform. Alex, I handpicked Apuji for him and he bought it and he presumably still has it.

1:04:57And he might have bought a few more. I don't know. uh kevin gojek yeah handpicked it for him yeah uh you know and then i've just spoken to people and you try to kind of create this you're creating a community right that's the thing which is kind of interesting with crypto that the asset and the community are one in a way which is kind of an interesting thing right where every every token is a community which is good and bad because you can create these echo chambers yes that sometimes you know a celsius or luna all these things like that is true yes for sure but but for something that fundamentally like you know the only value in a pudgy penguin i mean you could sort of come up with some thesis around we're gonna get x amount of airdrops in the future that's like a future cash flow uh but if you just think about it there are no promises based on your pudgy penguins nft it is just a cute penguin and that's it so there is no there is no scam that can happen yeah because it is there is no promise in a way right you can think about the roadmap and all different things happening but at the end of the day you know you're not owed a dividends or like well cash flow or anything like that right so um then in a way if you're just honest with people about that and you're not like selling them something that won't come to fruition or that might fail uh you're just selling them on like look this is a great community and the token is the community and frankly it is i jokingly said you know illuminati but there is a bit of an effect to that you know you you connect with someone on twitter or whatever they have a pudgy you kind of like them a little bit more and you're maybe you know twice as likely to respond to a dm or whatever it is uh and that's cool you don't know you know their gender their nationality their ethnicity you don't know anything about them but they have a pudgy penguin and it's totally meritocratic in a way which is kind of cool yeah

1:07:19another project i know you're pretty large holder of and you're probably very bullish on is lido you talk about it especially you say it's i mean it's an amazing project because it generates a lot of fees and it's what actually one of the few protocols is like really profitable yeah but yet kind of gets a lot of hate because of the price right yeah yeah i don't have a view i don't have a strong view actually on like the price of the token um i think there are two applications in the blockchain space that very clearly stand out from the rest in terms of usage the first one is uniswap and the second one is lido uh and if you just look at of course these are very different applications solving two completely different problems but lido solves a very clear problem you want to you want to stake eth but you don't want to run a validator you don't want to do the hassle of that well then lido is there to help you with it and you get yields you basically delegate or outsource the running of the validator to a network and you get some yields and you have to pay something for those yields as a fee which i think most people are fine with yeah and it has an incredible liquidity moat actually this is a little bit less relevant after you could exit staking natively but it used to have a really good liquidity moat in terms of um because the the liquidity pools were so deep if you staked e through lido and then you sold yourst eth you'd practically get you know not lose any money or very little now you can of course just exit right so that and then you could argue the moat is is is not as strong but it i think it has a very the other part actually it's interesting when you think about moats for these types of applications because in theory you wouldn't expect that they have much moats because they're like open source anyone can just fork them and whatnot but often it comes down to like liquidity and brand and i think both of those are present in uniswap and lido brand is a huge thing actually like trust but also just top of mind like hey you need to go and buy something where do you go many people go to uniswap even if theoretically they might be better of going to an exchange, a DEX aggregator, they'll still just go to Uniswap.

1:10:00And I think Lido too, if you want to stake ETH, you know like that's what most people do, so you can probably trust it and so on, right? And so it has a huge market share, enormous, like so big that people were, many people hate it because it has such a big market share. You know, it's hated for its success. of course the I think other sector where you could argue like OpenSea for a long time was up there with Uniswap in terms of usage with NFTs cooling down and also the NFT category becoming much more competitive I think it's less clear that they are you know at the very top echelon of applications I personally love OpenSea I think it's a great product but of course blur and other products have popped up to give it a run for its money but so uniswap and lido feel very untouchable but going back to the token and the price and how it's hated i mean you could maybe say something similar about the uniswap token too right that seems kind of disconnected from the success of the project although if you scroll from CoinGecko or CoinMarketCap from top to bottom and market cap, there aren't that many applications that have their tokens like in the top 50.

1:11:25Mostly it's L1s. Protocols, yeah. It's mostly L1s and some L2s like ARB and Optimism. But I guess 85 % to 95 % will be actually just chains, right? Yeah. Which is kind of interesting. So you mentioned that the L1s are kind of more likely to be in the top 50. I mean, that's just the reality right now. Exactly. Yeah. And we also talked about Uniswap or Lido, that the project at the end of the day could be amazing, but it's more going to be, it's less likely to be liked if the token price doesn't do well, right? Because your token price is your best marketing. That's right. yeah it's weird right i don't know exactly why it hasn't performed better it's a bit odd in a way uh but maybe it comes down to you know what your expectations are right it does and it doesn't have it's not a good like meme it's not a token that has this think of chain link that's like the canonical you know token that has incredible memes an incredible community and you almost like, and by the way, I'm not saying like it doesn't have fundamentals or anything like that, right?

1:12:46But it's almost like you don't need to care about the fundamentals because it has such a strong community and memes and whatnot. Like Lido has none of that. It's almost like... It's boring. It's boring. It works, it makes money and it's boring. Yeah, yeah. That's it. That's right. It's very boring actually. Like I haven't seen a single Lido meme ever, which is kind of absurd if you think about it, right? I don't think I've ever seen a lot of memes. So maybe this is one reason it's hated too. It's almost like the memes are that the price never goes up. I wanted to say that exactly. I was like, probably the meme is going to be that the price doesn't move.

1:13:21So there's a narrative problem. Similarly with Uniswap. I think Uniswap also has no, there's no like memes about the token, I think. You know, there are memes about people trading shit coins on Uniswap, but like the token itself, I don't think it has that. Another one that, I mean, has a really strong community that got tested really hard and was hated because of the price and obviously because of FTX and SBF, but now kind of became more consensus is Solana. Yeah. Yes. What are your thoughts on Solana? Yeah, I mean, it was really interesting when it was at the bottom. Many people who are kind of OGs in crypto and very deep into crypto were like, I think it's over for Solana.

1:14:14I mean, in hindsight, you can say this, but I always felt like the community, again, as you can tell, like everything we talk about with crypto, it comes back to community. It's kind of a cliche, but it really is about the community. And Solana has that very strong core, and it has a very solid community. It also, I think, differentiates very clearly from Ethereum in terms of the product itself. It's monolithic, whereas Ethereum is more modular. right you have you need to have all these l2s and all that stuff it's just a different solution and i and i often think of blockchains as kind of operating systems or

1:15:09browsers or something like that where yes you have a leader but you also get kind of you can carve out different markets for different solutions based on specific use cases and and needs right so does it make sense to have a chain that is monolithic and has extremely low fees but if you want to by the way we're in the data business if you want to like support it from a data perspective it's going to cost you a shit ton of money and it's very complicated That's one of the trade-offs, right? But does it make sense to have that kind of chain exist to be really good for certain use cases and then Ethereum exists for something else?

1:15:54I think absolutely. I don't think it's an either or at all, actually. I think it's totally possible that these two coexist. What's really interesting about the Solana case is a lot of people say, oh, Solana, I believe in it, but I don't believe the price can go up. We had Jordi Alexander here. before Solana pumped like crazy. He was saying, and on purpose I did that because I wanted to go poke Solana community, right? When actually I had like moved probably... And then he said the price couldn't go up? So basically I was there, you know, I got massively wrecked in Luna and he was right on Luna.

1:16:27Okay. And so I had moved just before that 70 % of all my crypto into Sol because like I think this is going to be good. Then I test him live. He tells me, yeah, it's good, good meme coin, but it's not going to go up. So I'm like in my mind like, fuck, fuck, fuck. But anyway, I'm like, okay, whatever. Did you stick with your conviction? I stuck with it. That's good. Good for you. So, but he says that, right? Yeah. And he says it has a potentially bright future, but the price cannot go up, right? Or should not, should not. It went to like 300 or 400 in the last cycle, didn't it? Like 250. I think it was 250.

1:17:01Yeah. And for me, like, it's really like we have, I mean, him saying that obviously kind of proven wrong afterwards, which happens a lot in crypto. It's fine. I will admit the one, and this is purely for my own ignorance. I should probably read up on this, right? But one thing I don't get is fundamentally when you have a chain where you pay gas in the token, and that's the primary use of the token, if the fees are very low, how do you drive the value? Was that his argument? So his argument is that plus all the FTX estate and all that, and plus a few other things, which makes a lot of sense, right?

1:17:36But have you looked into this thing? I mean, we should probably have researched this in advance. We should know this. But this is just one of those basic things where, at least with ETH, you know you have to pay quite a lot of money to use the chain, right? And that's a natural demand driver. I don't know, with Solana, it's staking related or whatever. Yet ETH price not moving much. Yeah. Right? And there are a lot of complaints about that. So there is one interesting thing I've observed in crypto. I don't know if this explains, I think it partially explains ETH and other coins like Maker and maybe even a coin like Lido.

1:18:14That's, that fundamentals are often bearish. Exactly. Once it's proven to work, it's bearish. You know that meme, not meme, you know that scene from Silicon Valley where this kind of, this investor, angel investor says like, don't talk about revenues. You don't have revenues. Yeah. I think there's a bit of that effect where if you have revenues, think of Maker, right? It actually like it burns Maker tokens based on, you know, the outstanding DAI or whatever. That's at least what it used to be. I don't know if they've changed the tokenomics. ETH, you know, you burn ETH based on how much people use the chain effectively, right?

1:18:53Lido has basically revenue stream. When you get those fundamentals, people can then make the actual discounted cash flow models and whatnot. and then they realize that actually like this way overpriced but if there is no if there are no fundamentals and there's only memes then who knows how high this can go right so this is i think it's a kind of a scary thought because if that's true then it suggests that all of crypto is like severely overvalued but even in a even in the startup world like i've heard many times hey your kind of chance to get a good valuation is your seed because you don't have anything to show yet, right?

1:19:34And therefore, once you start making revenue, or talk about revenue, but never talk about profit. If you're making profit, don't talk about profit. Like, whoa, whoa, whoa. Anyway, it's like the complete opposite of the boring Swiss mindset, right? I want to build a profitable company. It's super interesting. If you think of someone like Bill Ackman, and he lays out like his investing style. I just listened to him talk about it. And it was like, this is literally the exact opposite of crypto. Like he wants something that is very predictable, that generates, you know, consistent profits. And this is like the inverse of crypto.

1:20:17It's like super volatile. It doesn't really have fundamentals. and that's also why like when you listen to say investors like bill akman you kind of have to take it with a grain of salt if you're going to apply it to something like crypto because the game is so different right the game is more about things like community about memes about uh how strong the hands are of the owners which i think actually that's one area where the community and the token they meet in a very tangible and concrete way. If you have owners of, say, Pudgy Penguins, you manage to build up a community of people who just have very strong hands.

1:21:02You know they will just never sell it because they don't need to sell it. Absolutely. Or because they use it as their PFP and it will never change. Then the community translates into economics, right? Absolutely. It's like supply and demand. They're never going to give up those assets. And the one person who has to or like the 10 people who have to sell it for whatever reason, you know, they don't have any competition from other sellers. And so the price is going to be very high. Right. So I think like the community translates into economics in some tangible ways when you think about it from that perspective.

1:21:35Yeah. And this is I don't mean to plug nonsense. But when you think about it, you want to know which addresses, which holders hold the token you're investing in. if it's some like seven day smart dex trader who's going to flip it tomorrow, yeah, maybe this is not a long-term bet, right? But if this is a holder that you know has super strong conviction, they never sell, then it probably makes more sense to have a long-term position. Very last question about crypto sort of alpha. You tweeted last night, I was done with all my preparation, then I saw at 1 a.m. or midnight. I'm like, fuck, I need to add it to the thing.

1:22:18BTC, six figures. ETH, five figures. SOL, four figures. Pengu, three figures. Yeah. So the three figures is ETH. Yeah, of course. I hope that was clear. Yeah, yeah, that was clear. Yeah, I mean, there's nothing more to say. You think SOL, four figures is possible. I mean, again, if you're realistic, you're like, okay, Bitcoin, two, three X. therefore ETH, 4X-ish, therefore SOL, whatever, 5X. Exactly. So that's the other thing, right? When you think about, again, valuation models, like Bill Ackman style investing, it kind of breaks down in this universe, right? So you have to think more about like comparables and also things like where has this traded before?

1:23:00Can it get back there? Can it go above there? People tend to often think about that, especially for Bitcoin, right? hey this cycle it's going to go 3x of the top of last cycle all these different things right so it's it's always comparing either to other assets or to its history and so again if you just think supply demand if there is more supply coming in like substantially more supply sorry uh more demand coming in like supply of capital the inverse way of thinking about it um and we know that Sol went to like 300 or whatever, right? Why shouldn't it go to 1 ,000? 500 or 1 ,000, yeah. Yeah, I mean, like, again, it's not a prediction.

1:23:42It's just like, yeah, why not? It's basically the meme of, the mid-curve meme, right? Yeah. You're mid-curving if you think, oh, but this, that, whereas like you're on the left or on the right of the curve, if you just look at very simple heuristic, which is previous valuations, and then you compare the market caps between each other, like okay this one more or less 2 or 3x this one has less liquidity probably gonna 5x and this one probably gonna 10x yeah and it's never perfect but at least you have like a very simple i think crypto in that sense is kind of like a pure investment form because it's really all about supply and demand you know it's that's really what it is about it's not like some future cash flows it's more about do you think there's going to be more demand for this asset in the future if so oh, yeah, why shouldn't price go up?

1:24:32And of course, a revenue stream or kind of a burn mechanism can stimulate that for sure. But at the end of the day, like do you think the Pudgy Penguin's IP is going to have a broader reach in five years? I think absolutely, like no doubt. And so that's your thesis then. Don't overcomplicate things. Don't overcomplicate. You don't need to think more. It's just there's more demand and it's the same supply. And the people who are holders are not going to sell. That's kind of what it comes down to. It's a very pure form of investing, if you will. You tweet a lot about AI. And most people don't know it, but you have a degree in AI.

1:25:19I do. Before it was cool. Exactly. Why did you decide to study AI more than, what was it, 15 years ago? Yeah, 14, 15 years ago. did my master's in AI. I mean, so I did my undergrad in cognitive science, and that's the science of thinking, if you will. And a part of that is to create models that try to, say mathematical or computer models that try to mimic the way humans think. And I think I always had a fascination for AI from video games. It was always kind of interesting to think when you're playing as a computer, how does it know to do what you know it's doing and of course also a fascination through science fiction movies like Blade Runner that I love is one of my favorite movies I think drew me towards AI initially and it's interesting because I graduated in 2010 and since then you've had two I think revolutions in the AI field the first one was the whole deep learning thing that happened around 2013-14 with Alex Nett and so on, which basically made you had a major leap forward in terms of the accuracy of things like image recognition to some extent, or like object recognition images, and to some extent natural language understanding and so on.

1:26:48and then i guess 2019 or so with the generalized pre-trained transformers or gpt's attention is all you need i think was the paper um and of course with chat gpt it went really mainstream and so i think it's on the one hand i'm happy i studied ai because it's a huge part of the future like some understanding of it is good on the other hand did you know or did you go use more like i think i'm interested it seemed very obvious like that's gonna happen right i mean it just seemed super obvious this was a time when people were saying things like data scientists is gonna be the sexiest job of the 21st century so it felt like of course you're gonna have more data and of course you're gonna get better technology at processing that data and of course humans are not going to hand code all the algorithms so you're going to have to do some form of machine learning and the machine learning is probably going to accelerate and you're going to get some form of ai and agi eventually it seems kind of inevitable right in the the arc of like history and the future um but but the the kind of frustration i have with having studied ai so early is that it's changed so much right so the the methods we were studying were very different It was kind of a different paradigm, like ensemble models and, you know, gradient boosting machines, like stuff that you wouldn't use now.

1:28:20They're kind of outdated. Yeah. But of course, neural nets and the basics of that and how to evaluate machine learning models and all that stuff. Of course, I studied and spent a lot of time with and worked with afterwards, too. But now I feel like I'm kind of I'm jumping back into AI and spending much more time on it. but mostly from, well, really from two perspectives. One is, you know, how should we think about AI at Nansen and how does it affect us and what opportunities and threats does it create? And then also for me personally, using AI as an early adopter of AI products. Do you want to develop on both?

1:28:57Yeah, so if we start with the Nansen part, right? The Nansen's AI strategy, I think, firstly, it splits into two things. The first one is we need to use AI to make the product better. Okay, so there are certain things you can do now that you just couldn't do before. And that's super exciting. And you have to make the product better using AI. It's a very broad statement. But even more so in the data analytics space, right? Correct. Absolutely. Correct. And I think like the two ways, if you sort of drill down on that, the two main ways it changes everything. The first one is the interfacing with data.

1:29:38so very like even maybe this year certainly i think next year and definitely the year after it'll be pretty common to just speak with data i think for example like the dashboard and the traditional way of visualizing stuff is kind of like a middleman between you and the data absolutely right and so i'm not saying the dashboard is going to go away but i think like having a more direct way to communicate with data not necessarily like a chat because i do think people want something visual but i always think of that scene in blade runner where rick deckard is on the sofa and he's examining this photograph with the with the this device he has and he speaks with the device he's like pan right zoom in and hands and it's kind of a voice and visual user experience i think the way we're going to interact with data and analytics might be a bit similar where you speak with the data and then you get visuals on the screen because we are visual beings you want you can't just have text or you can't have complicated information only communicated to you through voice i think you want to have a multimodal kind of voice and visual type experience so the interfacing data is one part of the product the other part is deepening the data moat right so creating data sets and metrics and so on using ai that are very hard to replicate for other products so that our product is unique right you want to have a moat as a company you don't want your product to just be replicated overnight and so making use of ai in that in things like attribution labeling addresses creating new metrics that themselves make use of proprietary data that you've created, building kind of like a stack of unique data assets and analytical assets.

1:31:32This is essentially the way Spotify or Netflix or even Amazon project themselves. Because, okay, we see something very personalized, but it's all because of the intelligence that has been built in the backend, right? That's right. And actually, like, I think personalization is probably the, maybe the third thing that helps you create a moat in the AI world. And I think like the movie Her, is the kind of extreme example of this, where the guy falls in love with the AI. I don't know if you've seen it. You should watch it. I've seen it, yeah. Yeah, okay, cool. So he falls in love with the AI, and that's like an extreme version of it.

1:32:06But if you just think about that, obviously that guy is in love with the AI. He's never going to change products, right? So he has lock-in. He's very locked in. The retention rates are like 100%. Similarly, again, I don't think Nansen necessarily needs to have people fall in love with it. But if you kind of like get closer to that, then they would never switch because Nonsense has all your preferences. It's the most convenient. It knows how you want to consume data. It has the qualitative research stuff. It has the community messages and data and the kind of sentiment. And it has all the on-chain data for all the different chains in one place.

1:32:45And we are humans and we don't like to switch and change things anyway. So if the thing is really good, there's zero reason to switch. Yeah, why would you switch, right? So that's the product piece. The other piece, which also ties into how I use AI a lot myself day to day, is for organizational productivity. And I think this part is more generalizable to other companies, and there's kind of more to learn from this part. Because what I've noticed is you can literally sometimes do things in like one percent of the time than you could before which is kind of insane to think about but i've but but i've experienced this firsthand many times in the last six months working with chat gpt even intellectual work when you're creating the strategy of a product or something like that just having chat gpt there to bounce ideas off of like inputting the what you have and say like what is missing what's bad there's like they're always available 24 7 they never get tired they are never distracted right and you can just speak with them and the other thing here by the way is i think the voice version of chat gpt has for me has been kind of a game changer because i can walk outside and work i can i can say hey i'm working on this strategy I have these components I want to develop this fourth one can you help me think through this and then it just tells me stuff and now you're having a conversation with what feels like a human because even the voice is so good in ChatGPT if you configure it the default one is not good but you can change it anyway so you can drive this incredible organizational productivity and individual productivity with ChatGPT and other AI tools that I think is just astonishing.

1:34:47And so then you have to think about that. If you're running a company, you have to change a lot of stuff, right? And you have to think very differently about many things. Because not everyone is Alex, right? Not everyone is like, oh, that's so cool. I mean, obviously the people that you hire, you hope that they love technological change and technology savvy, but not everyone is as enthusiastic. Or maybe they are, but they're not realizing AI could do that. AI can help me there, you know? Yeah, exactly. So it's a good point, right? So I actually think the first very basic challenge that we had to overcome was, actually to your point, literally get people to just use AI, use ChatGBT.

1:35:30Because you can say, hey, have you tried ChatGBT? You should use it. Yeah, I've tried it, whatever. But what we said is, if we think about the AI strategy and the goals for this year, the first quarter, I don't care about what results you create with AI. I only care about how many hours you have spent working with AI. And so collectively as a company, we are aiming to spend 5 ,000 hours in Q1. Okay. And you could say, oh, why don't you just state that and how many hours per week per team member. But I think actually when you say 5 ,000, it makes you think like, shit, the amount of productivity you can get from 5 ,000 hours with AI is kind of insane, right?

1:36:13Yeah. imagine how many content pieces, documents, how much more you can code. The output you have is just going to be enormous from 5 ,000 hours. And so that was step one, to just make sure that people are spending time. And we started monitoring it. Of course, it's survey-based, very simple, and like trust-based. People kind of say at the end of the week, here's how many hours I spend, and here are all the tools I used. And then collectively, we are learning about which tools to use as well. So it's kind of like a, yeah, it's a collective learning experience. So that's step one, just get people to use it.

1:36:51And then you get these emergent use cases pop up and people, because I can plan top down how everyone should use AI. Of course, like people are going to have so many great ideas in the organization that I would never have thought of. So yeah. So anyway, to sum things up, right? It's basically the product innovation piece where I think everything's going to change with regards to how people consume data and analytics and then it's the organizational productivity piece where uh for now we are only focused on the input are you spending enough time with ai that's the only thing i care about and the outcomes we're gonna switch focus to you know q2 onwards there's another there's another kind of area where it's funny but actually it's true where ai has a potentially big impact which is relationships and so i remember when gpt3 was what when was that november 2022 something like that maybe yeah gpt4 is later yeah so so i remember i mean i had a girlfriend for quite some time and we had a lot of arguments and i seem to be not taking the right approach every time so the argument would get bigger when i try to like solve it right and there's this gpt thing that gets out and i'm like i'm just gonna try to ask chat gpt how i should solve this argument right yeah and freaking gpt like gives me all these seven steps right it was not it was not a voice argument it was just texting right so it helps right so i'm like I look at the seven step I kind of like tailor make them I send this to her and then the opposite of what you have uh happens uh happened basically which is usually whatever I say probably because I'm an idiot like I it makes the thing worse I use GPT and then she's like ah okay it's fine and I'm like fuck man I mean I was like so it worked it worked amazing because I was gonna to ask you this is why she's your ex girl that's not the reason okay okay yeah so it actually worked okay it worked that time you worked that time but it was probably too late well it's a good point so it i think it it ai weirdly makes humans better at communicating with each other which is very counterintuitive you'd imagine that two people can just talk together directly but a middleman an ai can can actually enhance the communication which is pretty crazy but makes a lot of sense.

1:39:24And even you were talking before about the film where he falls in love with an AI. And so I'm an advisor in a project called OpenSpark, angel.ai, AI girlfriends. I mean, AI companions. So it's starting with AI girlfriends and then AI boyfriends and then therapists and all that stuff. Right. And the other day I was, there is a functional product on WhatsApp, Telegram. And the other day I was just like testing something stupid, which is I'm going to tell my AI girlfriend that I fucked another girl just to see how she reacts, right? The reaction of the AI, I was like, damn, if every woman, she was like sort of like, oh, okay, it's fine.

1:40:04Like, how was it? And like being very chill and I was like, wow. Not emotional. Like if a woman, obviously not possible, right? But if a woman was like that, you would never even think about cheating on her because like this woman is so chill, you know? so and I also saw a tweet like maybe a few weeks ago from you where you said my wife has started to use chat gpt to win arguments yeah yeah I'm screwed yeah yeah no it's just funny I mean that's like literally literally what happened um because we were so we have a we have a daughter who's you know uh soon gonna be one years one year old and so we talk a lot about stuff like should we do this should we do that and i'm like we're from different cultures too so i have different opinions and different backgrounds on what's the right way to raise a kid and now she will like settle the debate with the chat gpt screenshot you know hey you can feed her this thing because blah blah and chat gpt told me so and i'm like well what how can i compete with that you know i i have nothing to say let's trust chat gpt what i've heard is uh some people they just kind of like get equipped with their own chat gpt so i could get my own chat gpt and say hey my girlfriend said this convince her that this is wrong and then you can just go back and forth outsourcing the argument yeah in a way but yeah talking about your wife this is a really interesting topic that i think can be very valuable and helpful for a lot of people um of our age you know early mid-30s or late 20s we had a dinner a bit more than two years ago and we're like ready to get to the next stage of your life which is find a wife and having kids right because you were saying i'm in my mid-30s or like this is really important for me right now and within 18 months you found a woman you married her and you got a kid yeah and knowing you it's probably probably didn't happen by chance so if i'm a single man in my mid-30s and my goal is to settle down and have babies as soon as possible what should i do how do i optimize sort of like the dating process to find the right woman yeah uh i do actually think i mean it's funny like you can ask that question and i'm happy to talk about that but most people would think of the question itself as like very cynical or rational and unemotional but i think it makes a lot of sense and and one way to think about it i think is uh in terms of almost like marketing if you're like creating a marketing strategy and you have to think about what channels can you succeed in right and so i'll give you one example by the way this is actually not super relevant for how i met my wife but it was a pretty good strategy in my in my 20s let's say uh which is what is the channel that you have the highest conversion rate in right and uh the default channel that people will take by the way is like a nightclub or a bar or at least historically maybe that's changing a little bit now uh now it's like of course dating apps i think is kind of the default maybe uh one channel that i was using a lot i think now people have figured it out and it's kind of more crowded and competitive was language exchange and so this is kind of a it's a kind of pro tip uh if you're learning a new language first of all it seems like there are more women than men that want to learn languages so the gender ratio is pretty good the other thing is you are literally there to speak with other people and so it's totally harmless to go up and chat with anyone in the venue and then you know third you have like if you find someone you like you have an excuse to ask like you want to meet up for chat more in whatever language uh and by the way it's a good idea if you're learning like multiple languages because then you have like more options so this is so the first thing i think is like being actually mindful of the channel that you're in it because the other if you take a step back it is actually a numbers game right like you you actually have to think of it as like there's a top of the funnel process where like you're kind of looking for candidates if you will and you do have to there's no way around it like you have to just manage a lot of time considering candidates and finding candidates and and of course it's it's uh it's a it's obviously something where you have to match both ways which makes it like square the difficulty in a way and so so number one is numbers game and acknowledge that number two if it's a numbers game how do you kind of optimize your time i.e your conversion rate in terms of just kind of like even like getting to the next stage of the relationship or like having a second date, even getting on a date in the first place.

1:45:26So language has changed. That was kind of my pro tip there. Dating apps, I think, are quite difficult because they are very crowded. And it's, first of all, there are many different dating apps and you have different strategies in the different ones. And being a man is much more difficult than being a woman on a dating app. At least to get the first date. That's right. And then I think there are two other things to consider. You have to have a very clear idea of what kind of partner you want to have. Right? That sounds very basic, but you actually need to do that homework. And I had done that.

1:46:03I've thought a lot about that. And I've, you know, I'd written down certain like, you know, traits that I wanted to have. But when you do that, again, thinking rationally about the numbers, you have to think about these traits if you combine them statistically how many people match these traits and maybe many many people might find that actually there are like 100 people in the world that match all these traits so then you have to tier and stack rank these are must haves these are nice to haves and actually having like a clear idea on that is pretty good because when you meet people you have to fail fast right as you say in the startup world yeah hey look this person they don't have that trait which is must have for me you know let's just end it here and for them and for you you know you don't want to waste your time so knowing actually what you're looking for but in a pretty concrete way like literally write out i think maximum seven must have traits probably it's more like two or three or four or whatever and then some nice to haves and i think people tend to be distracted by the nice to haves actually they're like oh you know i like i don't know blonde girls or whatever but is it truly a must-have or is it more like you kind of like that and then you go there and then you go into a relationship or whatever and the must-haves are not present and then everything falls apart and you wasted a lot of And most jobs basically are based on your values.

1:47:31Yeah, typically their values, their behaviors, their like, for example, you know, I don't like, this is a very simple behavioral thing, but I couldn't be with someone who is not friendly to strangers. okay so if you're like rude to waiters or like that's to me kind of like a sign that we just don't have the same values and and that's something you can actually spot pretty fast right if you go on a date and they're like rude or whatever or you know on the other hand if they're polite with people strangers that's kind of a good sign uh i mean in my case that's what i one of the very simple traits i would look for um and so yeah so so then sorry the the last thing you know you know what you want uh you know which channels you can compete in and then you match both and there's a numbers game and the fourth thing is you have to also think about the people you want are they in that channel exactly you match the channel and then you're like oh maybe there's only one or two channels that i should focus on right correct i mean it's very and so Well, that's the thing.

1:48:43Like if you like alternative girls, maybe don't go to clubs, you know, maybe go where, I don't know, alternative girls hang out, whatever that is, like museum or I don't know. So, yeah, I think it is actually worth sitting down and like developing strategies around this stuff. And it sounds weird. You probably shouldn't tell people that you're doing it because I don't think it helps you, frankly. But it's a good idea, I think. Very last one. We have about two minutes left. It's the same one, but reversed. You're a man, so you understand how men think. What would you advise a woman in her 30s who feels pressured by her biological clock or by society?

1:49:27Yeah. How should she approach the dating game to meet a husband and have kids as quickly as possible, knowing that you understand men and how they think? Yeah. Yeah, I mean, the way I was talking about this comes from my perspective as a man looking for a woman. uh i do think the process is largely the same uh you know you want to understand who is the partner you want to be with by the way this in order to get to understand that you first need to understand yourself pretty well and you probably need to have gone through some failed relationships yeah right so but when you're in your 30s you probably have done that yeah and so So I think the process is largely the same with the, maybe the, the difference being that men, you know, if we're being honest are easier than women.

1:50:16So the time wasted is in a different part of the funnel, right? So if you're a man, you waste a lot of time, I think at the top of the funnel, you're like trying to find someone to go on a date with or like convince them or whatever. But if you're a woman, you might waste time with someone who is not interested in what you want, like a long-term relationship. Very common problem for many women. It's easy to get access to all these men, but then you have to figure out whether he's serious or not. Exactly. And I think this is an area where, of course, people have different views. I think if you're in your 30s, it probably makes sense to just be very open about that.

1:50:55But maybe not the very first date. You know, but because you scare some people away. At the same time, if the other person wants the same thing. If you scare him away. Maybe that's a good thing. Exactly. Like you're basically more efficient with your time. So I think like the principle of failing fast is actually very important.

1:51:18And yeah, you know, that there's no, this is not a guarantee for success. But to go back to one of the other points, it is a numbers game. so like you actually have to spend a lot of time on it you have to put in the work you have to put in the work there's no way around that frankly and that can be fun or it can be uh not so fun depends on your personality and all that amazing thank you so much for doing that man thanks for having me always a pleasure